Abstract

Neurofibromatosis type 1 (NF1) is the most common inherited cancer predisposition syndrome and is characterized by the development of benign and malignant peripheral nerve sheath tumors. While plexiform neurofibromas (PNs) are frequent and often indolent, a subset progresses through atypical neurofibromas (ANs) to malignant peripheral nerve sheath tumors (MPNSTs), which are aggressive, therapy-resistant, and the leading cause of-cancer-associated mortality in NF1 patients. Early and reliable detection of malignant transformation and residual disease remains a clinical challenge due to limitations in current imaging, biopsy, and tumor heterogeneity in NF1 patients. In the first study, we used ultra-low-pass whole-genome sequencing (ULP-WGS) and cfDNA fragment size profiling to distinguish MPNSTs from PNs in 53 individuals. Tumor fraction and fragmentation features classified MPNSTs with 86% accuracy pretreatment and 89% during serial monitoring. Shorter cfDNA fragments and elevated tumor fractions were associated with MPNSTs, and dynamic changes in cfDNA tracked with tumor burden and treatment response. In the second study, we applied whole-genome cfDNA sequencing to a larger cohort of 101 NF1 patients and 21 controls. We analyzed fragmentomic features including bin-wise size ratios, end motifs, and nonnegative matrix factorization (NMF) to further resolve the tumor continuum. Fragmentomic signatures improved detection sensitivity for MPNSTs and enabled differentiation of ANs from both PNs and MPNSTs, achieving AUCs of 0.75–0.77 for these transitions. In at least one case, cfDNA analysis correctly identified malignancy where tissue biopsy misclassified the tumor. In the third and final study, we developed a targeted cfDNA sequencing assay in 82 individuals using a custom panel enriched for genes recurrently altered in MPNSTs. The integrated model, combining SNVs, CNAs, and structural variants (SVs), achieved a cross-validated AUC of 0.904 and outperformed tumor fraction estimates derived from genome-wide analysis alone. Serial sampling demonstrated the clinical utility of this approach for real-time monitoring: ctDNA detected a TP53 deletion prior to radiographic relapse in one patient, and complete ctDNA clearance was observed during remission in another. Together, these findings support a multimodal cfDNA-based strategy that integrates global fragmentomic and targeted mutational analyses for accurate, noninvasive detection and monitoring of malignant transformation in NF1. Given its scalability and precision, this approach may also inform liquid biopsy development in other solid tumors and heritable cancer syndromes.

Committee Chair

Angela Hirbe

Committee Members

Aadel Chaudhuri; Brian Van Tine; Christopher Maher; Kelly Bolton

Degree

Doctor of Philosophy (PhD)

Author's Department

Biology & Biomedical Sciences (Molecular Cell Biology)

Author's School

Graduate School of Arts and Sciences

Document Type

Dissertation

Date of Award

11-6-2025

Language

English (en)

Author's ORCID

0009-0002-7785-0962

Download

Available for download on Friday, November 05, 2027

Included in

Biology Commons

Share

COinS